Prostate cancer metastasis to bone is associated with significant morbidity and is not curable with current treatment regimens. Since the pathophysiology of prostate cancer (PC) metastasis to bone is poorly understood it has been difficult to identify appropriate treatment modalities. In previous studies we have noted that PC cells that induce osteoblastic lesions express different cytokines than PC cells associated with osteolytic lesions. This suggests that different signaling pathways are responsible for the formation of osteoblastic and osteolytic lesions. In addition, our data also suggests that osteoclast activity may not be necessary to establish tumor cell intravasation into bone. Our goal is to enhance our understanding of the biological interaction between prostate cancer cells and bone in order to eventually identify potential treatment modalities that can limit or inhibit the metastatic process. We propose the hypothesis that PC cell expression of RANK-ligand and/or BMPs directly control the phenotype of metastatic bone lesions. Therefore, we will determine the influence of RANKL and BMP inhibition on the formation of human PC bone lesions in a tibial injection model in SCID mice. In Specific Aim 1 we will determine the role of RANK inhibition on the formation of osteoblastic, osteolytic, and mixed lesions, in bone by administering RANKFc to the SCID mice. The research proposed in Specific Aim 2 will test the hypothesis that BMPs secreted by tumor cells play a role in the bone formation seen in osteoblastic lesions. Prostate cancer cell lines will be stably transfected with a retroviral vector expressing the BMP antagonist Noggin. In Specific Aim 3, we will determine the influence of both BMP inhibition and RANK blockade on the formation of osteoblastic and osteolytic lesions in bone. The Noggin expressing PC cells generated in Aim 2 will be injected into the tibias of SCID mice that are treated with RANKFc. These experiments will either identify the dominant metastatic phenotype or elucidate a novel combination therapy to effectively treat tumor metastasis to bone. This proposed research should enable to us to identify the role that RANKL and BMPs play on the development of tumor induced osteolytic, osteoblastic or mixed lesions in bone and hopefully will take us a step closer in identifying potential treatment mordalities to prevent or limit metastatic prostate cancer to bone.